The immobilization of enzymes on a matrix is a well-established approach to improve their stability and (re)usability for industrial applications. While many different concepts for enzyme immobilization have been developed, in vivo immobilization (i.e., heterologous expression of a target protein and subsequent coupling to biogenic carrier materials within living cells) is still challenging. Moreover, online monitoring of the in vivo immobilization process is not yet possible. Here, we present an alternative method for in vivo immobilization using split fluorescence proteins, such as split GFP, as a universal linker and simultaneously as a sensor for immobilization. For this, the GFP sequence is split into two non-fluorescent fragments GFP1-10 and GFP11 that have the ability to self-assemble to generate a specific green fluorescent signal. We used split GFP as a linker to immobilize target proteins on the surface of heterologous polyhydroxybutyrate (PHB) granules in Escherichia coli. While GFP11 is fused to the PHB synthase PhaC as an anchor protein, the target protein is linked to GFP1-10, so that successful immobilization is accompanied by a green fluorescence signal. As a proof-of-concept, the red fluorescent protein mCherry, the ester hydrolase LipD from Alcanivorax borkumensis and the squalene synthase from Methylococcus capsulatus were selected as target proteins. To verify the immobilization process in E. coli cells, we used fluorescence spectrometry, microscopy, and fluorescence lifetime imaging. LipD activity was analyzed in vitro using 4-nitrophenyl butyrate as substrate and HPLC analyses were performed to determine the amount of squalene produced in vivo by the squalene synthase.

Overall, we were able to show that the split GFP system can be used as a combined linker-sensor system for the in vivo protein immobilization on the PHB surface without affecting the function of the target proteins. Thus, we have developed a new system that enables an easy to detect decoration of the PHB surface and may therefore represent a promising approach for future applications.